R and geodata, part 3

Andreja Radović, PhD

25th of January 2018

Need for geo analyses?

Examples from my research

Site selection

Example: Site selection

Agrale site selection

Agrale: site selection

  • Most recent orto-photo images
  • Image classification and interpretation
  • vegetation plots – predefined abandonment class
  • 26 habitat classes
  • habitats of interest (BLFO | CGR | HGR | PA | SHR | WL | OFO | HD) == proxy for woody structure at plot!
  • 4 abandonment classes * 8 habitat classes * 5 replicates * 4 countries* - random selection from defined strata *height layers

Agrale after site selection

agrale

Agrale after site selection

agrale 2

Agrale - birds

  • Systematic sampling on Passerine bird communities
  • Standard point-count method
  • circular sampling plots 100m radius

agrale 3

Agrale - birds/landscape

agrale 5

Example: Selection of research sites NIP/WB

NIP

Example: Selection of research sites NIP/WB

NIP2

Example: Selection of research sites NIP/WB

NIP3

Example: data collection, geocoding

  • storks via Google maps API from R

  • schools in Croatia via HERE Nokia API from R

White stork habitat suitability

  • Combination of regression techniques and geostatistics
  • Detect suitable yet still not occuoied areas (to be important in predicted climat changes)
  • Predict extected nest densities at 1km scale

Geocoding

Example: data collection

Black sork, Slovenia

  • Remote sensing (satelite tracking)
  • Animal movement
  • Habitat usage
  • Lidar point cloud analysis

Black stork, satelite

Satelite tracking of animals

Black stork, satelite

After: analyses according research question

After: analyses according research question

After:

After:

Example: Species distribution modelling - RK

Ecological modelling

Example: Habitat change detection

CLC change

Example: Habitat change detection

CLC change

Example: Habitat change detection

CLC change

Example: forestry practice

Influence of forestry practices

Example: forestry practice

Data used

Influence of forestry practices 4

Example: forestry practice

Aquila pomarina

Influence of forestry practices 3

Example: forestry practice

Haliaeetus albicilla

Influence of forestry practices 4

Example: forestry practice

Influence of forestry practices 5

Example: Climate/habitat characteristics

  • Climate variable
  • Habitat classification
  • Gower algorithm (klastering – data mining)

Climate habitat

Example: Climate/habitat characteristics

Climate habitat 2

Example Presence of suitable feeding habitats

Platalea leucorodia

Example Presence of suitable feeding habitats

Platalea leucorodia 2

Automatisation of spatial modelling

Automatisation of modelling procedure

Automatisation of spatial modelling

Automatisation of modelling procedure

Automatisation of spatial modelling

Automatisation of modelling procedure

Usage of biased sampling data

Problem in most of DB in Croatia

[UseR 2017! Conference] (https://www.user2017.brussels/uploads/radovic_user2017.pdf)

Data bases

Usage of biased sampling data

Data bases 2

Usage of biased sampling data

Data bases 3

Usage of biased sampling data

Data bases 4

More data mining

  • morphological characteristics of plants due to climate differences

Morphology climate

More data mining

Morphology climate 2

More data mining

Morphology climate 3

Finally

Valentino Perović diplomski FER

What is geo data?

What is geo data?

Geo data - topology

Altogether - data

Various sources:

Thank you for your attention!